Applying Deep Learning at Cloud Scale, with Microsoft R Server & Azure Data Lake

Applying Deep Learning at Cloud Scale, w/ Microsoft R Server & Azure Data Lake

  • Figure 4: Generating training data in parallel using Microsoft R Server.
  • We present the final tagged test image in Figure 8 where cars and boats are labeled with red and green bounding boxes respectively; you can also download the image .
  • Each worker node returns a labelled list of moving window tile coordinates, which is then used to label the final test image in MRS running on HDInsight Spark edge node.
  • We compress 2.3 million training images from 8.9GB of raw PNG images to 5.1GB with im2rec binary in 10 minutes for optimal training performance.
  • MXNet DNN model training using NVIDIA Tesla K80 GPU using Microsoft R Server (MRS).

This post is by Max Kaznady, Data Scientist, Miguel Fierro, Data Scientist, Richin Jain, Solution Architect, T. J. Hazen, Principal Data Scientist Manager, and Tao Wu, Principal Data Scientist Manager, all at Microsoft.

Continue reading “Applying Deep Learning at Cloud Scale, with Microsoft R Server & Azure Data Lake”

Marr Revisited: 2D-3D Alignment via Surface Normal Prediction

Marr Revisited: 2D-3D Alignment via 
Surface Normal Prediction  #deeplearning #cmu #cvpr16

  • Our two-stream network allows us to retrieve CAD models that better match the style and pose of a depicted object compared with baseline approaches.
  • We introduce a skip-network model built on the pre-trained Oxford VGG convolutional neural network (CNN) for surface normal prediction.
  • Our model achieves state-of-the-art accuracy on the NYUv2 RGB-D dataset for surface normal prediction, and recovers fine object detail compared to previous methods.
  • Marr Revisited: 2D-3D Model Alignment via Surface Normal Prediction
  • When using the predicted surface normals, our two-stream network matches prior work using surface normals computed from RGB-D images on the task of pose prediction, and achieves state of the art when using RGB-D input.


We introduce an approach that leverages surface normal predictions, along with appearance cues, to retrieve 3D models for objects depicted in 2D still images from a large CAD object library. Critical to the success of our approach is the ability to recover accurate surface normals for objects in the depicted scene. We introduce a skip-network model built on the pre-trained Oxford VGG convolutional neural network (CNN) for surface normal prediction. Our model achieves state-of-the-art accuracy on the NYUv2 RGB-D dataset for surface normal prediction, and recovers fine object detail compared to previous methods. Furthermore, we develop a two-stream network over the input image and predicted surface normals that jointly learns pose and style for CAD model retrieval. When using the predicted surface normals, our two-stream network matches prior work using surface normals computed from RGB-D images on the task of pose prediction, and achieves state of the art when using RGB-D input. Finally, our two-stream network allows us to retrieve CAD models that better match the style and pose of a depicted object compared with baseline approaches.

Continue reading “Marr Revisited: 2D-3D Alignment via Surface Normal Prediction”